This invention generally relates to methods for detection and confidence evaluation, and more particularly to methods for detection and confidence evaluation of vital sign signal.
With increased attention to health care, detection requirements for vital sign are increased. And noncontact vital sign sensing system has recently become a focal point because it has no restriction on installation site and has a longer detection distance than contact type vital sign sensing system.
The noncontact vital sign sensing system radiates wireless signals toward a part of human body and receives reflected signals from the part of human body. If the body part has a movement relative to the noncontact vital sign sensing system, the body movement may generate Doppler Effect on the wireless signals so that the reflected signals may contain Doppler shift components caused by the body movement. The noncontact vital sign sensing system can measure the body movement by demodulating the reflected signals, and the measured body movement is regarded as vital sign signals of human body while the body movement is caused by vital sign.
However, undesired movements, such as movements of other objects or other parts of human body, and noise signals may interfere with the noncontact vital sign sensing system to lead the reflected signals contain Doppler shift components caused by the undesired movements. Accordingly, it is difficult to identify the movements caused by vital signs of human body in the reflected signals and is possible to identify the undesired movements as the movements caused by vital signs.
The object of the present invention is to compute a confidence of a vital sign signal according a eigenvalue of the vital sign signal and a reference value by using a compute module, so that user can determine whether measured displacement information is the vital sign signal.
A method for detecting vital sign signal of the present invention comprises following steps: capturing an image by using an image capture module; determining whether there is a living body captured in the image by using a variation detection module; changing a beam direction of a vital sign sensing system toward the living body; detecting a vital sign signal of the living body by using the vital sign sensing system; receiving the vital sign signal from the vital sign sensing system and computing a eigenvalue of the vital sign signal by using a compute module; and evaluating a confidence of the vital sign signal according to the eigenvalue of the vital sign signal by using the compute module.
A method for evaluating confidence of vital sign signal of the present invention comprises following steps: detecting a vital sign signal of a living body by using a vital sign sensing system; receiving the vital sign signal from the vital sign sensing system and computing a eigenvalue of the vital sign signal by using a compute module; and evaluating a confidence of the vital sign signal according to the eigenvalue of the vital sign signal by using the compute module.
In the present invention, the variation detection module determines whether there is a living body captured in the image and the compute module computes the confidence of the vital sign signal to confirm whether the living body determined by the variation detection module is a living body having vital signs, so that the living body captured in the image is confirmed twice. By the vital sign sensing system, the vital signs of the living body captured in the image can be measured automatically.
With reference to
With reference to
With reference to
With reference to
Further, the setting value may be a temperature range based on types of the target subject while the image P is a dynamic thermal image captured by a thermal imaging camera. For example, the temperature range may be between 32° C. and 40° C. when the target subject is human being, and the variation detection module 200 may determine the pixels belong to the living body O if the temperatures of the pixels in the image P are within the temperature range.
With reference to
The reflected signal Rs received by the antenna 320 is transmitted and injected to the self-injection-locked oscillator 310 to make the self-injection-locked oscillator 310 operate in a self-injection-locked state and generate a self-injection-locked signal SIL. The reflected signal Rs contains the Doppler shifts caused by the displacement of the living body O and the frequency variation of the self-injection-locked signal SIL is proportional to the Doppler shifts caused by the displacement of the living body O, so that the displacement information of the living body O can be obtained by demodulating the self-injection-locked signal SIL.
As shown in
Positions of the pixels determined as the living body O are identified according to an image coordinate of the image P, but the image coordinate is unavailable for identifying the position of the living body O when changing the beam direction B of the vital sign sensing system 300. As a result, the method 10 preferably further includes a step of transforming the image coordinate into a real coordinate by the variation detection module 200 before changing the beam direction B of the vital sign sensing system 300. The beam direction B of the vital sign sensing system 300 is able to be directed toward the living body O in accordance with the real coordinate.
With reference to
The variation detection module 200 in the step 12 only utilizes the image P to determine a moving object or an object having higher temperature as the living body O, but can't confirm whether the determined living body O is a living body having life characteristics. That is to say, the vital sign signal VS detected by the vital sign sensing system 300 may be displacement of a non-living body or noise from the space S, not vital signs of living body. Hence, the method 10 preferably further includes steps 15 and 16 used to determine whether the vital sign signal VS detected by the vital sign sensing system 300 is from vital signs of living body.
With reference to
While the demodulator 330 is a I/Q demodulator and the signal demodulated by the demodulator 330 has no DC offset component, the amplitude for a period of time is represented by
A=√{square root over (I2+Q2)}
where I and Q are in-phase component and quadrature component of the vital sign signal VS, respectively.
While the demodulator 330 is a I/Q demodulator and the signal demodulated by the demodulator 330 has no DC offset component, different calculation formulas of the phase for a period of time are given by
one of the above calculation formulas of the phase for a period of time is selected according to circuit architectures and initial phase of the I/Q demodulator.
If the signal demodulated by the demodulator 330 contains DC offset component, the amplitude for a period of time is represented by
A=√{square root over ((I−DCI)2+(Q−DCQ)2)}
where I is in-phase component of the vital sign signal VS and DCI is DC offset of in-phase component of the vital sign signal VS.
If the signal demodulated by the demodulator 330 contains DC offset component, different calculation formulas of the phase for a period of time are represented by
one of the above calculation formulas of the phase for a period of time is selected according to circuit architectures and initial phase of the I/Q demodulator.
With reference to
The calculation formula of the confidence is given as follows:
where C is the confidence of the vital sign signal VS and At is the reference value of the amplitude. The confidence is expressed in linear form so that the higher the calculated confidence value, the higher credibility the vital sign signal VS has.
In other embodiment, the calculation formula of the confidence may be represented as follows:
where W is a constant and the confidence is expressed in exponential form.
Likewise, the reference value may be a median or a constant of the phase of the reference vital sign signal for a period of time when the eigenvalue of the vital sign signal VS is the phase.
In this embodiment, not only determine whether any living body O is captured in the image P by using the variation detection module 200, but also determine whether the living body O determined by the variation detection module 200 is a living body having vital signs according to the confidence obtained by the compute module 400, so that the living body O captured in the image P is confirmed twice. Accordingly, vital signs of living bodies in the space S can be detected automatically.
Not a single information of the vital sign signal VS, the eigenvalue in other embodiment may be a combination of multiple information, such as a combination of single information (amplitude or phase) of the time domain signal at multiple different time points, a combination of different information (amplitude and phase) of the time domain signal at same time point, a combination of different information of the time domain signal at multiple different time points, a combination of single information (amplitude or phase) of the frequency domain signal at multiple different frequencies, a combination of different information (amplitude and phase) of the frequency domain signal at same frequency, a combination of different information of the frequency domain signal at multiple different frequencies, or a combination of information of the time domain signal and the frequency domain signal. Like the eigenvalue, the reference value also may be an amplitude, a phase, a combination thereof or a weighting combination thereof of the time or frequency domain reference vital sign for a period of time or a time point.
The scope of the present invention is only limited by the following claims. Any alternation and modification without departing from the scope and spirit of the present invention will become apparent to those skilled in the art.
Number | Date | Country | Kind |
---|---|---|---|
107124756 | Jul 2018 | TW | national |